Metastatic cutaneous melanoma is resistant to conventional therapies, resulting in poor prognoses. About 60% of these melanomas harbor the BrafV600E mutation, and most are responsive to the recently approved therapies that target the mutant Braf protein. However, disease generally recurs within 12-15 months. Additional promising results that demonstrate the possibility of durable responses in a subset of patients have recently been achieved with immune checkpoint modulation therapy that targets PD1/PDL1 or CTLA4 or the combination, and further immunomodulation therapies are being developed. However, it is not clear why some patients respond so well, while others don't respond at all. In addition, it will be critical to determine dosing regimens, effective combinations of multiple immunotherapies and their combination with cancer cells or other components of the cancer stroma. Using the established melanoma GDA models described in project ZIA BC 011462 of this report, we have initiated studies for all objectives. Thus far, we have shown that about 30 % of HGF/MET-driven primary tumors are resolved upon treatment with anti-CTLA4, which is similar to results in patient trials. Tumors in the existing Braf model do not respond. FACS and immunohistochemistry studies to immunophenotype tumor-bearing animals are in progress. All methodologies have been established. In preliminary antigenicity studies, some, but not all HGF/MET-driven tumors register as antigenic, while Braf tumors are not. This is consistent with the hypothesis that positive therapeutic responses are associated with increased antigenicity. However, the experiment is being repeated in parallel with treatment of GDAs established using the same tumor tissue used for antigenicity evaluation. We have shown that anti-CTLA4 treatment combined with small molecule MEK inhibition can improve responses compared to either agent alone, but only when agents are delivered in certain dosing regimens. Further assessment of dosing regimens is in progress. In collaboration with Medimmune and Dr. Merlino, we have begun experiments to evaluate the efficacy of ant-PDL1 in the HGF/MET GDA model. Experiments are in progress and will follow the same standard operating procedures and immune system analyses as previously applied to anti-CTLA4 evaluation. Similar experiments are planned to test anti-PD1 once the mouse specific antibody has been validated by Medimmune. including models in which tumors express luciferase. In addition, primary Braf driven and primary and metastatic HGF/MET/Braf driven models have been established. All models have been validated for responses to targeted therapies (vemarafinib in Braf and crizotinib in HGF/MET models).